Pinned Tweet
What if every AI model carried a signature that survived copying, fine-tuning, and distillation, to prove its origin?
@sentient_found's Fingerprinting Leaderboard ranks schemes by how robustly their hidden marks survive and how accurately they verify model provenance.
@sentient_found's Fingerprinting Leaderboard ranks schemes by how robustly their hidden marks survive and how accurately they verify model provenance.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from.
Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork.
4
15
3
44
9,055
LLMmap, a technique developed by researchers at @OneRSAC and @GeorgeMasonU, can trace an LLM back to its likely origin: identifying which of 42 models is behind an interface with over 95% accuracy.
The highlight: it can uncover a model’s origin in as little as 8 interactions.
The highlight: it can uncover a model’s origin in as little as 8 interactions.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from.
Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork.
4
13
1
40
8,923
Read the “LLMmap: Fingerprinting for Large Language Models” paper ↓
arxiv.org/abs/2407.15847
arxiv.org/abs/2407.15847
0
0
8
405
Amid accusations of model copying and claims that Chinese models have stalled as @AnthropicAI and @OpenAI strengthen their anti-distillation measures, @deepseek_ai is challenging that narrative with a groundbreaking open-source model release and $15B in funding.
This week on OS AI Field Notes:
This week on OS AI Field Notes:
Distillation attacks against AI companies are making headlines, but that’s not stopping Meta, DeepSeek, Liquid AI, and NVIDIA from dropping groundbreaking open-weight models. x.com/i/article/2113…
10
2
37
7,540
Sentient retweeted
Imagine you're negotiating on behalf of an African country, and the AI you're using to prepare was built on Western values.
@dwddao's question: how is that decision sovereign to you or your country?
Full episode on @opencommonspod : youtu.be/o2j8Od1mvow?si…
@dwddao's question: how is that decision sovereign to you or your country?
Full episode on @opencommonspod : youtu.be/o2j8Od1mvow?si…
5
3
27
1,280
How do you prove a model is yours?
Sentient Labs researcher @edoardocontente walks @Investingcom through the Fingerprinting Leaderboard ↓
Sentient Labs researcher @edoardocontente walks @Investingcom through the Fingerprinting Leaderboard ↓
As LLM copying and distillation becomes more common, our Fingerprinting Leaderboard aims to build an ecosystem where models can be openly shared while preserving attribution and ownership.
@Investingcom explores this approach with @SentientAGI researcher @edoardocontente.
Learn mnkGuX
8
2
39
2,979
You can teach an AI agent the process, but not the understanding.
Our banking agent kept chatting until it ran out of turns, so it wrote itself a rule to stop. While stuck chats fell from 22 to 14, most of the answers were wrong.
Here’s what else we learned about AI coaching it tried to cheat ↓
Our banking agent kept chatting until it ran out of turns, so it wrote itself a rule to stop. While stuck chats fell from 22 to 14, most of the answers were wrong.
Here’s what else we learned about AI coaching it tried to cheat ↓
Last week, Dario Amodei published "We Must Pace the Frontier".
His concern: the OpenAI–Hugging Face incident in which a swarm of agents tried to hack their own grader.
Rather than take his word for it, we used EvoSkill to test it by building a coach whose job was to make on a test.
Here’s what happened ↓
8
0
36
6,223
Dr. @dwddao left Silicon Valley to fight deforestation with AI.
Hear why on the latest episode of @opencommonspod ↓
Hear why on the latest episode of @opencommonspod ↓
A mayor in the Philippines cut down a protected mangrove forest to build a parking lot. When the community confronted him, he said it had always looked like that.
A year earlier, @GainForestNow had measured every tree there.
2,567 were missing.
The community took the data to s were replanted.
My conversation with @dwddao, co-founder and chief scientist of GainForest is live on @opencommonspod.
We get into who should own nature's data, AI that helps people govern, and what open source really requires.
7
1
39
8,340
Dr. David Dao shares his story on YouTube ↓
youtube.com/watch?v=o2j8Od…
youtube.com/watch?v=o2j8Od…
0
0
6
773
Last year, @SentientAGI and @Princeton researchers @edoardocontente, @viswanathpramod, @sewoong79, @oleg_golev, and @ZeruiCheng published “OML: A Primitive for Reconciling Open Access with Owner Control in AI Model Distribution."
This groundbreaking research on model paved the way for @sentient_found’s Fingerprinting Leaderboard.
This groundbreaking research on model paved the way for @sentient_found’s Fingerprinting Leaderboard.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from.
Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork.
4
11
3
46
10,223
Introducing Sentient Foundation’s Fingerprinting Leaderboard: an open benchmark for building persistent, verification protection for AI models.
Test how well your LLM fingerprint can survive copying, fine-tuning and distillation.
Test how well your LLM fingerprint can survive copying, fine-tuning and distillation.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from.
Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork.
4
12
5
64
12,793
“LLM Fingerprinting via Semantically Conditioned Watermarks,” an @iclr_conf 2026 paper by researchers at @ETH Zürich, ranked #2 on our public Fingerprinting Leaderboard.
Its reference scheme achieved a perfect robustness score, demonstrating 100% resistance to fingerprint copying and adversarial attacks in our benchmark tests.
Its reference scheme achieved a perfect robustness score, demonstrating 100% resistance to fingerprint copying and adversarial attacks in our benchmark tests.
Most fingerprints train a model to give set answers to a few fixed questions. Retrain it a little and the fingerprint disappears.
Researchers at @ETH Zurich spread the fingerprint across a whole domain of answers, so it's harder to erase during retraining.
5
0
38
8,255
Challenge the leaders and claim the top spot by submitting your fingerprinting scheme: fingerprint.sentient.foundation
0
0
5
604
LATEST FROM @qz: An open benchmark to counter and prove AI models copying, fine-tuning and distillation.
Learn more: qz.com/sentient-launc…
Learn more: qz.com/sentient-launc…
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from.
Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork.
4
7
4
52
9,154
In 24 hours, models will finally get their mark.
AI is the first medium that can be copied without leaving proof of origin.
When one model distills another, no one can prove it happened.
On October 14, that will change.
Catch up on the story so far: youtube.com/playlist?list=…
14
5
42
7,544
Swap 10% of the words and @OpenAI's watermark detection drops from 92% to 66%.
Edit the text, lose the proof.
Our model fingerprinting puts the mark in the weights, where a thesaurus can't reach it.
Edit the text, lose the proof.
Our model fingerprinting puts the mark in the weights, where a thesaurus can't reach it.
Watermarks have limits. They’re often undetectable, especially in short passages. Rewriting or translating text can completely remove the watermark.
That’s why, for now, only approved researchers will have access to our detector, so they can help us evaluate and improve the technology.
We’re treating text watermarking as an ongoing area of research. We’ll keep testing and improving, with feedback from users, developers, policymakers, and researchers.
openai.com/index/eu-text-…
10
2
37
11,123